Ion movement across excitable membranes takes place is discrete "channels", endowed with ion selective properties. The mechanism(s) of ionselectivity in such "channels" are only partly understood, especially in mixtures of electrolytes, i.e. under physiological conditions. Electrostatic interactions between "channels" and ions have a significant role in determining this selectivity. However, other factors, possibly structural changes in the "channels", play a major but unknown role. The classical "carrier" concept of transport across biological membranes is found to be inconsistent with experimental data. New models are being developed, based on polypeptide interactions within lipid bilayers, but at present few experimental data are available concerning such interactions. Studies on gramicidin induced ion transport across artificial biomolecular lipid membranes offer unique opportunities to study mechanisms of ion transport through narrow "channels" as well as molecular interactions of polypeptides in lipid bilayers. The objectives of this proposal are: (1) to study ion transport through single gramicidin channels; (2) to study the ion selectivity properties of these chaanels; (3) to examine whether structural changes of the channels may play a significant role for the ion selectivity, and; (4) to study the molecular interactions of gramicidin A, and some natural and synthetic analogues, in artificial lipid bilayers.